On Tue, 2011-08-09 at 10:33 -0700, Tobias Gogolin wrote: > I don't like to believe that that is the best 'servo' conversion one could > make adding a readout, one that would try faster pulses and lock... > So if I can work with these motors, who can recommend a good high voltage > power supply and microstepping driver for these? > > Nobody has ever implemented a closed loop stepper controller in EMC? > > Thanks guys!
As has been indicated in other replies, a stepper or servo system won't work well if the motors or drivers are not up to the task. No amount of a feedback command will compensate for a weak drive system. So, I would think getting a good understanding of the load requirements should be done first. So far, my machines have come with motors and drivers, so I haven't had to work this out. I haven't tried this (and hope to someday), but in the past, I ran across a thread that suggested using a pulley fixed to a hand wheel or screw shaft(if your machine has them), then tying a weight and string to the pulley such that gravity can drive the screw. One adds weight to the pulley to the point that the axis moves at the desired rate and/or acceleration. The pulley radius and mass of the weight would give the required torque. This, plus some math, might be enough to evaluate motor data (most likely Watts and torque) to decide on the proper motor size, then size (most likely Watts again) the drive current and voltage to match the motor. Drive voltage is the major determinant of motor speed. Drive current is the major determinant of torque. Also the motor:screw ratio can be adjusted to get more torque or speed, and may allow one to live with an existing drive system by sacrificing the maximum axis speed down to the maximum cutting speed (in other words, sacrifice rapids to get enough cutting feed rate torque). It can be hard to determine how hard one can drive a motor, but I tend to use motor temperature. My guess is that one can drive the motor to the point of making it barely touchable, although high heat and flux can ruin motor magnets. For steppers, I have found the maximum temperature when the motors are at idle (but powered up). This drives one set of coils at 100% duty and all of the power is converted to heat, with none converted to motion (, except if the driver has an idle current feature). Cheap drivers won't have any or a decent current limiter, so they can be easy to burn out. A .1 or.01 Ohm, multi-watt resistor in series with the driver and in parallel with a voltmeter can be used to measure the driver current. Heat sinking the driver chip, FETS or anything that gets hot, can help. So, there are a number of things that could be looked at to find a solution, but the goal should be to size the system to the load, plus a bit of safety margin. If there is enough margin on a stepper system it will never skip a step and pretty much "just work". As for encoders on a stepper, it might be handy for catching a following error and stopping the job before ruining a part, or might be needed for an unusual application, but for the most part ( in my opinion) is not needed. -- Kirk Wallace http://www.wallacecompany.com/machine_shop/ http://www.wallacecompany.com/E45/index.html California, USA ------------------------------------------------------------------------------ uberSVN's rich system and user administration capabilities and model configuration take the hassle out of deploying and managing Subversion and the tools developers use with it. Learn more about uberSVN and get a free download at: http://p.sf.net/sfu/wandisco-dev2dev _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
